"Epitope discovery" is a new way to identify antigenic fragments of pathogen proteins. Starting with a library of tens or hundreds of millions of random fragments of pathogen proteins, anti-pathogen antibodies are used to select fragments that bind particularly tightly to subspecificities within the antibody population. Selection is accomplished with simple microbiological methods by means of phage display technology. The selected peptides, having won a rigorous competition among all the structures in the initial library, have strong credentials as candidate components of synthetic or recombinant peptide vaccines. But are these peptides good "immunogenic mimics"? That is, are they able on their own to induce antibodies that cross-react with the pathogen itself--as they must if they are to protect against disease? This key question will be addressed using bacteriophage T4 as a model "pathogen" (T4 is readily and safely produced in large amounts, allowing cross-reactions to be quantified directly by simple immunochemical techniques). From the results we will learn which intrinsic properties of peptides can be used to predict outstanding immunogenic mimicry a priori, without having to assess it directly in living subjects--a difficult task in the context of actual diseases. Immunogenic mimicry is necessary but not sufficient for vaccine performance, however, since not all pathogen-reactive antibodies actually protect against disease. Using a second model system--the cattle disease caused by the tick-borne, malaria-like parasite Babesia bovis--we will show how antigenic peptides can be screened in vitro for the likelihood of protective value. Taken together, the constellation of innovations to be explored--isolating particularly promising peptides via epitope discovery, assessing them for properties that correlate with superior immunogenic mimicry, and screening them in vitro for the capacity to engender protective effects--promise to deliver vaccine candidates with excellent prospects for efficacy, before a single trial in a living subject must be undertaken.